Rapid and accurate detection and identification of pathogens is crucial for the effective management and treatment of infectious disease. While molecular diagnostic techniques including PCR are a promising approach for rapid and sensitive detection of pathogens, they suffer from a number of technical drawbacks which have slowed their acceptance in the clinic. Microfluidic lab-on-a-chip technologies are well-poised to overcome many of these limitations by providing vastly increased levels of automation, throughput and reliability while reducing assay costs and instrument sizes. Our long term goal is to develop a highly-flexible lab-on-a-chip platform for detection of pathogens by real-time PCR based on our innovative nanoliter droplet-based "digital microfluidic" concept. Ultimately, we imagine that this platform would enable a new generation of highly-automated handheld instruments capable of providing fast, accurate, and multiplexed molecular test results for the diagnosis of infectious disease. Towards this goal, in phase I, we will demonstrate the feasibility of performing parallel automated real-time PCR reactions on our platform. We will develop a prototype chip capable of automatically setting-up nanodroplet reactions in a fully-programmable and combinatorial fashion. We will further demonstrate that thin-film heaters can be integrated directly into the chip to perform the thermocycling operations. In later phases the detector and control electronics would also be miniaturized or integrated onto the chip. Additionally, in phase II we propose to apply our prototype device to the problem of Candida detection in extremely low-birth-weight infants and rigorously validate our technology through participation in a multi-center clinical study in collaboration with Duke University. [unreadable] [unreadable] [unreadable]